A Matlab-based toolbox for characterizing behavior of rodents engaged in string-pulling

Abstract

<jats:p>String-pulling by rodents is a behavior in which animals make rhythmical body, head, and bilateral forearm as well as skilled hand movements to spontaneously reel in a string. Typical analysis includes kinematic assessment of hand movements done by manually annotating frames. Here, we describe a Matlab-based software that allows whole-body motion characterization using optical flow estimation, descriptive statistics, principal component, and independent component analyses as well as temporal measures of Fano factor, entropy, and Higuchi fractal dimension. Based on image-segmentation and heuristic algorithms for object tracking, the software also allows tracking of body, ears, nose, and forehands for estimation of kinematic parameters such as body length, body angle, head roll, head yaw, head pitch, and path and speed of hand movements. The utility of the task and software is demonstrated by characterizing postural and hand kinematic differences in string-pulling behavior of two strains of mice, C57BL/6 and Swiss Webster.</jats:p>

Deep learning-based behavioral profiling of rodent stroke recovery

“…Consequently, deep learning-based quantitative analysis has been successfully performed in various behavioral set-ups often on rodents for data-driven unbiased behavior evaluation [ 22 , 24 , 25 , 27 , 32 ]. Some data-driven approaches also used acute brain injuries and neurological disorders to reliably predict the extent of injury [ 38 , 44 ].…”

“…This function computes the Euclidean error between human-annotated ground truth data and the labels predicted by DLC averaged over the hand locations and test images. During training, a score-map is generated for all keypoints up to 17 pixels (≈0.45 cm, distance threshold) away from the ground truth per body part, representing the probability that a body part is at a particular pixel [ 27 ].…”

“…However, these technologies offer little methodological transparency, are not affordable for many laboratories [ 16 ], and are often designed to study pre-specified modules within one particular paradigm (e.g., the Morris water maze or the open field test) rather than discover new behavioral patterns. Although the concept of quantitative behavioral analysis has already been implemented in some cases before [ 17 ], the introduction of machine learning algorithms has recently reached various sectors of life and provided a new set of tools ideally suited for behavior analysis [ 18 – 27 ]. These algorithms, referred to as deep learning models, offer user-defined feature tracking with greater flexibility, as well as reduced software and hardware acquisition costs [ 28 ].…”

Inertial Measurement of Head Tilt in Rodents: Principles and Applications to Vestibular Research

“…In the context of experimentally-induced vestibular lesions in rodents, head tilt has traditionally been rated on a discrete scale by an observer, offering a qualitative estimation rather than a quantitative measurement [ 26 , 29 , 30 , 31 , 32 , 33 , 34 , 35 ]. In non-vestibular rodent research, approximations of head tilt angles have occasionally been calculated from photographs or single video recordings [ 36 , 37 ]. Recent efforts in 3D videography have yielded methods for estimating head tilt in rodents [ 19 , 38 , 39 ], but their accuracy has not been assessed.…”

A Primer on Motion Capture with Deep Learning: Principles, Pitfalls, and Perspectives

“…(A) In marker-based tracking, prior to performing an experiment, special measures have to be taken regarding hardware and preparation of the subject (images adapted fromInayat et al [2020] andMaceira-Elvira et al [2019]; IMU stands for inertial measurement unit).…”